(635f) Developing a Dynamic (PERFUSION BASED) Multicellular Pancreatic Cancer MODEL

INTRODUCTION: Recent advances in tissue engineering have led to the development of scaffold assisted 3D tumour models with better niche mimicking capability in comparison to traditional 2D systems. Additionally, they are easier and faster to use as compared to animal models. Their ability to better recapitulate the tumour niche is attributed to their tunable mechanical properties and structural integrity along with better cell- cell, cell-Extracellular Matrix (ECM) interactions [1]. We have previously reported that poly urethane (PU) based scaffolds can be used as a robust system for the development of robust mono-cellular (cancer cells only) pancreatic cancer model [2]. Furthermore, we have recently reported that our developed PDAC model is appropriate for short and long term chemo-radiotherapy screening [3]. However, in addition to cancer cells and matrix proteins, the tumour niche consists of different cell types, including stellate cells and endothelial cells all contributing to tumour formation, cancer metastasis as well as its response and resistance to treatment. Stellate cells in particular have shown to produce high amounts of ECM, leading to fibrosis-desmoplasia in pancreatic cancer. Thus, recent studies have been focused on the generation of multicellular models of PDAC, primarily spheroid based, which are of limited mechanical integrity [4]. Our group has very recently developed a novel hybrid scaffold assisted multicellular model with cancer cells, stellate cells and endothelial cells [5]. Our scaffold assisted multicellular model captures different levels of desmoplasia, which is a hallmark of pancreatic cancer, along with long term viability of all different cell types [5]. However, in vivo cellular niches have additional complexity associated with flow through vascular networks and in the form of interstitial fluid flow which is not recapitulated within a static system. Additionally, mimicking the circulation of immune cells also require a dynamic flow within in vitro 3D tumour models. However, very few studies exist till date to assess the effects of dynamic condition of monocellular of multicellular models of PDAC and they are mostly spheroid based and for relatively short time period [ 6,7].

The aim of this work was to build up on our novel scaffold based multicellular model and develop a long term (> 4 weeks) dynamic perfusion (bioreactor based)pancreatic cancer model, mimicking various physiologically relevant vasculature/interstitial flow levels.

METHODS: PU scaffolds were prepared using the Thermal Induced Phase Separation (TIPS) method. Absorption based surface modification of the scaffolds enabled ECM matrix remodelling with fibronectin and collagen type I [2,3]. Thereafter, hybrid scaffolds with different zonal ECM coatings were fabricated as described in [5], to account for ECM optimisation for different cell compartments. More specifically, a zonal structure with (i) endothelial and stellate cells on the outer side of the scaffold coated with collagen and (ii) pancreatic cancer cells in the inner scaffold coated with fibronectin was designed [5]. The hybrid multicellular scaffolds were then placed in a dynamic (perfusion) bioreactor. Different flow rates in the range of 0.5-5 ml/min were evaluated. Various in situ assays for monitoring the cell viability, spatial organisation and ECM production were carried out. More specifically, immunofluorescent assays and visualisation with CLSM, SEM. were carried out at specific time points throughout the cultures for different flow rates.

RESULTS: We have successfully established a zonal multicellular 3D model showing extensive desmoplastic reaction in the presence of stellate cells and cellular migration, mimicking key in vivo characteristics of pancreatic cancer. Addition of shear stress and fluid flow within a perfusion bioreactor resulted in discernible changes within the model in comparison to static culture in terms of cellular growth, matrix production and migration.

DISCUSSION & CONCLUSIONS: Our data show, for the first time, the feasibility of PU scaffolds to support a zonal multicellular pancreatic tumour niche growth along with the possibility for a robust ECM mimicry and recapitulation of fibrosis/desmoplasia. We also highlight the importance of fluid flow/ shear stress on in vitro cancer model. Different flow rates affect different scaffold compartments/cell types of the tumour microenvironment. Overall, our developed novel advanced pancreatic cancer model is a high throughput tool that can be used for animal free personalized studies and treatment screening of pancreatic cancer.

ACKNOWLEDGMENT: Financial support was received from the Department of Chemical and Process Engineering at the University of Surrey, Impact Acceleration Grant (IAA-KN9149C) from University of Surrey, IAA–EPSRC Grant (RN0281J) and the Royal Society. P.G was supported by Commonwealth Rutherford Post-Doctoral Fellowship. E.V. is thankful to the Royal Academy of Engineering for an Industrial Fellowship.

REFERENCE:

[1] Totti, et al. DDT.2017; 4(22).

[2] Totti, et al. RSC Advances.2018; 8(37).

[3] Gupta, et al. RSC Advances.2019; 9 (71).

[4] Lazzari, et al. Acta Biomaterialia.2018; 78.

[5] Gupta, et al. Frontiers in Bioengineering & Biotechnology.2020; 8:290. doi: 10.3389/fbioe.2020.00290

[6] Tai, et al. Phytomedicine. 2014; 21(4).

[7] Brancato, et al. Acta Biomaterialia. 2017; 49.